TCNeKP: A Novel Deep Learning Architecture for Enzyme Catalytic Activity Prediction

Accurate prediction of enzyme kinetic parameters (Kcat and Km) is crucial for enzyme rational design and engineering research. Based on a heterogeneous data set encompassing 17,893 Kcat and 24,585 Km records across 8911 enzyme sequences from 7 EC classes and 5023 substrates, we introduce novel TCNeKP models for predicting Kcat and Km values. Herein, enzymes' sequences were autoembedded and processed by a temporal convolutional network (TCN) module to extract the key features of catalytic and binding residues frequently located far apart in the primary sequences; substrates were encoded by a pretrained SMILES-Transformer language model; and catalytic conditions (pH and temperature) were encoded via radial basis function (RBF). The fused features were then fed into a fully connected network for single-task prediction of Kcat and Km. Results demonstrate that TCNeKP-Kcat and TCNeKP-Km models achieve robust performance across wild-type and mutant enzymes from 7 EC classes, outperforming state-of-the-art MPEK, UniKP, and DLKcat models (Table S3). Leveraging a cross-task dynamic parameter-sharing module with attention mechanism, we further developed a multitask TCNeKP model that achieves the highest R2 values among the benchmark models for both Kcat (0.677) and Km (0.657) prediction. These findings indicate that collaborative learning between Kcat and Km prediction tasks enhances feature extraction for enzyme–substrate binding and catalysis, thereby significantly enhancing the predictive performance.